Neuroscience 266 (2014) 38–46

BEHAVIORAL MODEL OF ITCH, ALLOKNESIS, PAIN AND ALLODYNIAIN THE LOWER HINDLIMB AND CORRELATIVE RESPONSESOF LUMBAR DORSAL HORN NEURONS IN THE MOUSE

T. AKIYAMA, M. NAGAMINE, M. I. CARSTENS ANDE. CARSTENS *

University of California, Davis, Department of Neurobiology,

Physiology & Behavior, 1 Shields Avenue, Davis, CA 95616, USA

Abstract—We have further developed a behavioral model of

itch and pain in the lower hindlimb (calf) originally reported

by LaMotte et al. (2011) that allows comparisons with

responses of lumbar dorsal horn neurons to pruritic and

noxious stimuli. Intradermal (id) microinjection of the prurit-

ogens histamine, SLIGRL-NH2 (agonist of PAR-2 and

MrgprC11) and chloroquine (agonist of MrgprA3) into the

calf of the lower limb elicited significant biting and a small

amount of licking directed to the injection site, over a 30-

min time course. Following id injection of histamine,

low-threshold mechanical stimuli reliably elicited discrete

episodes of biting (alloknesis) over a longer time course;

significantly less alloknesis was observed following id injec-

tion of SLIGRL-NH2. Capsaicin injections elicited licking but

little biting. Following id injection of capsaicin, low-thresh-

old mechanical stimuli elicited discrete hindlimb flinches

(allodynia) over a prolonged (>2 h) time course. In single-

unit recordings from superficial lumbar dorsal horn

neurons, low-threshold mechanically evoked responses

were significantly enhanced, accompanied by receptive field

expansion, following id injection of histamine in histamine-

responsive neurons. This was not observed in histamine-

insensitive neurons, or following id injection of saline or

SLIGRL-NH2, regardless of whether the latter activated the

neuron or not. These results suggest that itch-responsive

neurons are selectively sensitized by histamine but not SLI-

GRL-NH2 to account for alloknesis. The presently described

‘‘calf’’ model appears to distinguish between itch- and pain-

related behavioral responses, and provides a basis to inves-

tigate lumbar spinal neural mechanisms underlying itch,

alloknesis, pain and allodynia. � 2014 Published by Elsevier

Ltd. on behalf of IBRO.

Key words: itch, alloknesis, pain, allodynia, scratching, dor-

sal horn neuron.

INTRODUCTION

Chronic itch frequently accompanies many

dermatological and systemic diseases (Ikoma et al.,

http://dx.doi.org/10.1016/j.neuroscience.2014.02.0050306-4522/� 2014 Published by Elsevier Ltd. on behalf of IBRO.

*Corresponding author. Tel: +1-530-752-7767 (lab); fax: +1-530-752-5582.

E-mail address: eecarstens@ucdavis.edu (E. Carstens).Abbreviations: id, intradermal; WDR, wide dynamic range; HT, high-threshold; 5-HT, 5-hydroxytryptamine.

38

2006). The associated scratching exacerbates the skin

inflammation, leading to a vicious itch-scratch cycle

(Wahlgren, 1999) that reduces the quality of life

(Hundley et al., 2006). A potential mechanism is

sensitization of itch-signaling pathways, manifested by

spontaneous itch and scratching, hyperknesis

(enhanced itch), and alloknesis (touch-evoked itch)

(Hosogi et al., 2006; Ikoma et al., 2006; for recent

review, see Akiyama and Carstens, 2013). Itch can be

elicited by innocuous touching of normal skin around a

site of histamine-evoked itch (Bickford, 1937; Graham

et al., 1951; Simone et al., 1991; Heyer et al., 1997). In

chronic itch patients low-threshold mechanical stimuli,

such as clothes contacting the skin, can initiate the

itch-scratch cycle (Bendsoe et al., 1987; Wahlgren,

1991; Heyer and Hornstein, 1999; Ricci et al., 2006;

Mason, 2008).

To develop mechanism-based treatments for itch,

animal models are required. Scratching behavior in

rodents is commonly used to assess itch (Carstens and

Kuraishi, 2004). This traditionally involves counting

hindlimb scratch bouts directed toward a pruritogen

injection in rostral back skin. A drawback is that

hindlimb scratching does not discriminate between itch

and pain since it is the only available response. We

developed a novel model of alloknesis in which touch

near a site of intradermal (id) injection of histamine and

certain other pruritogens, or at the edge of an area of

dry skin, reliably elicited hindlimb scratches (Akiyama

et al., 2012a). In a recent variant, id pruritogen injection

in the rodent cheek elicited primarily hindlimb

scratching, while algogens elicited mainly forelimb

wiping (Shimada and LaMotte, 2008). Scratching was

reduced by systemic administration of l-opioidantagonists whereas wiping was reduced by systemic

administration of morphine (Akiyama et al., 2010a;

Spradley et al., 2012b). Furthermore, this model allows

direct comparisons of behavior with pruritogen- and

algogen-evoked responses of first- and second-order

trigeminal neurons (Akiyama et al., 2010b; Klein et al.,

2011a). For decades, however, much information about

itch and pain has come from recordings of lumbar spinal

neurons with input from the hindlimb. Moreover, it is

important to recognize differences in trigeminal vs.

spinal processing of itch and pain (e.g., Spradley et al.,

2012b). Thus, a behavioral model that distinguishes

between itch and pain from the lower hindlimb is

desirable.

T. Akiyama et al. / Neuroscience 266 (2014) 38–46 39

Hindpaw injection of 5-hydroxytryptamine (5-HT)

elicited naloxone-sensitive paw-biting accompanied by

licking, while hindpaw injection of algogens only elicited

paw-licking (Hagiwara et al., 1999). Mice with chronic

dry hindlimb skin exhibited spontaneous paw-biting

(Nojima et al., 2004) that was sensitive to naltrexone

but not morphine (Akiyama et al., 2010c). Lumbar

dorsal horn neurons ipsilateral to the dry skin-treated

hindpaw exhibited elevated spontaneous activity

(Akiyama et al., 2011a,b). A minor drawback of the

hindpaw behavioral model is that weight-bearing and

locomotion may interfere with itch-related behaviors. It

was recently reported that id injection of histamine into

the mouse calf elicited biting whereas injection of

capsaicin elicited licking (LaMotte et al., 2011). In the

present study, we have further developed and validated

this calf model. We additionally developed a novel

model of alloknesis and allodynia, whereby low-

threshold mechanical stimuli reliably elicit biting or

hindlimb flinches, respectively, following id injection of

histamine or capsaicin. Finally, in electrophysiological

experiments we tested whether id injection of histamine

enhances the mechanosensitivity of pruritogen-

responsive lumbar superficial dorsal horn as a possible

mechanism underlying alloknesis.

EXPERIMENTAL PROCEDURES

Experiments were conducted using adult C57BL/6 mice

(Harlan, Oxnard CA, USA) (19–32 g body weight) under

a protocol approved by the UC Davis Animal Care and

Use Committee.

Behavior

The fur on the calf was shaved and mice were habituated

to a Plexiglas recording arena with a transparent cover

one week prior to testing. On the test day, the animal

was restrained by hand and the skin on one hind limb

was mildly stretched. An id microinjection of 10 ll wasmade in the calf of one of the following: vehicle (isotonic

saline), histamine (50 lg; Sigma–Aldrich, St. Louis, MO,

USA), SLIGRL-NH2 (50 lg; Quality Controlled

Biochemicals, Hopkinton, MA, USA), chloroquine (50 lg;Sigma), capsaicin (10 lg; Sigma), or 7% Tween-80

(vehicle for capsaicin). The calf id microinjection was

made using a 30-G needle attached to a Hamilton

microsyringe by PE-50 tubing. Immediately following the

id microinjection, mice were placed into a clear glass

arena containing mirrors set at angles to allow multiple

views of the animal that were captured with a high-

definition videocamera. Investigators left the room during

videotaping. The videocamera was set at high definition

and slow motion capture modes to facilitate the

assessment of biting and licking behaviors directed to

the injected calf in a frame-by-frame video playback

conducted offline by two investigators blinded as to the

experimental treatment. The duration of biting, licking

and flinching episodes was timed at 5-min intervals over

the 50-min recording period. A bite is defined as direct

contact of the incisors with calf skin. Biting was

accompanied by relatively high-frequency low-excursion

head movements which were used as an adjunct

measure. The duration of each biting action was timed

with an original program which allows us to check the

movie frame by frame, and individual bite durations were

summed to provide the cumulative biting time. The

cumulative duration of licking, defined as direct contact

of the calf skin by tongue protrusion, was also

measured. Licks were accompanied by head

movements that were of lower frequency and larger

excursion compared to those associated with biting, and

were also used as an adjunct measure of licking.

In a separate experiment, the animal was habituated

to a recording arena for 1 h and then tested for

alloknesis and allodynia at 5-min intervals, starting 5 min

post-histamine or capsaicin injection. Alloknesis and

allodynia were assessed as follows. At 5-min intervals,

the mouse received three separate innocuous

mechanical stimuli delivered using a von Frey filament

(bending force: 0.7 mN). Stimuli were delivered 2 mm or

further distal to the edge of the visible bleb formed by

the id injection of histamine or other agents. The 0.7 mN

von Frey filament was selected because it never elicited

behaviors (i.e. biting, licking and flinching) in naı̈ve mice,

and was the minimum strength to elicit biting when

delivered to skin surrounding the site of histamine

injection. The mice were videotaped with a high-speed

recording feature that turns 3 s of video into 12 s of slow

motion video. The presence or absence of a positive

response, i.e., biting, licking and flinching directed to the

site of mechanical stimulation, was confirmed by video

playback. The behavior score was the total number of

positive responses elicited by the three stimuli, i.e., 0, 1,

2 or 3. The sequence was repeated at 5-min intervals

out to 60 min post-injection, and again at 90 and

120 min post-injection time points. In experiments with

histamine and capsaicin, an overall behavior score per

60 min and 120 min, respectively, was calculated as the

sum of individual behavior scores.

Electrophysiological recording

The mouse was anesthetized with sodium pentobarbital

(60 mg/kg i.p.) and prepared for single-unit recording

from the lumbar spinal cord as previously detailed

(Akiyama et al., 2009b). A tungsten microelectrode was

driven into the superficial lumbar dorsal horn and single

extracellularly recorded units were isolated using

mechanical touch and pressure stimuli delivered to the

ipsilateral hindpaw. In a few experiments we recorded

from units that responded to mechanical stimulation of

the ipsilateral calf. However, we decided to focus on

units with hindpaw receptive fields because data from

such units may be compared with data from our

previous electrophysiological studies (Akiyama et al.,

2009a,b, 2011a,b), and it was more difficult to

accurately map receptive field dynamics on the calf

compared to hindpaw. Recording depths were restricted

to <300 lm below the surface as in our previous study.

Unit activity was amplified, digitized and displayed on

computer using a Powerlab (AD Instruments, Colorado

Springs, CO, USA) interface. Once a mechanosensitive

unit was isolated, we tested the unit responsiveness to

40 T. Akiyama et al. / Neuroscience 266 (2014) 38–46

light brushing with a cotton wisp, followed by pinching

using forceps. Units were classified as wide dynamic

range (WDR) if they responded at higher firing rate to

pinch than light touch. They were classified as

high-threshold (HT) if they responded to pinch but not

light touch. The areas of mechanosensitive receptive

fields were mapped by determining the perimeter along

which application of the same von Frey filament used in

behavioral studies (0.7 mN) evoked a response on 50%

of application trials. Since some units did not respond, a

series of von Frey filaments of progressively stronger

bending forces ranging up to 760 mN was used to map

the receptive field area. Each unit was retested with the

0.7 mN von Frey filament at least three times to

establish a baseline response level, which for some

units was zero. Five minutes later, either histamine,

SLIGRL-NH2 (both 50 lg/ll in saline), or saline, was

microinjected id (1 ll volume) within the

mechanosensitive receptive field. Unit receptive field

areas were redetermined 5 min postinjection, and at 5-

min intervals thereafter out to 30 min postinjection, and

again 45-min and 60 min postinjection. Unit responses

to application of the 0.7 mN von Frey filament at the

same sites tested preinjection were also determined at

the same time intervals. At the conclusion of the

recording, an electrolytic lesion was made. The spinal

cord was postfixed in 10% buffered formalin and cut in

50 lm frozen sections to identify the lesion sites under

the light microscope.

Fig. 1. Time course of biting and licking elicited by histamine, SLIGRL and ca

vs. time following id injection of histamine (50 lg/10 ll) in the calf skin of the

(50 lg/10 ll). (C) Capsaicin (10 lg/10 ll). (D) Saline. (E) 7% Tween 80. (F) M

Significant difference between bite and lick durations (p< 0.05, paired t-tes

Unit activity was usually quantified as number of

action potentials/s or min. Responses to von Frey

stimuli were summed over a 20-s period and responses

to histamine over successive 60-s periods. A positive

response was defined as a 30% or greater increase in

the total number of action potentials per 20 or 60 s post-

stimulus compared to the same time interval before the

stimulus. Averaged responses to von Frey stimuli before

and after application of the pruritogen (or saline) were

compared by paired t-test with P< 0.05 set as

significant.

RESULTS

Chemically evoked biting and licking behavior

Following id injection of histamine into the calf, mice

exhibited biting directed toward the injection site, with

significantly less licking (Fig. 1A, F). Biting peaked

within the first 10 min post-injection and declined over

the ensuing 30 min (Fig. 1A). Biting was characterized

by contact of the incisors with the skin in a fairly high-

frequency and low-excursion motion of the head. In

contrast, licking was characterized by repeated

protrusions of the tongue toward the skin over a longer

excursion and lower frequency that could be readily

distinguished from biting. The cumulative duration of

biting was significantly greater following id injection of

histamine compared to saline controls (p< 0.05,

unpaired t-test). A similar time course of biting, with

psaicin. (A) Cumulative duration of biting (j) or licking (4) behaviors

hindlimb. Error bars: SEM (n= 6). (B) Same as in A for SLIGRL-NH2

ean cumulative duration of biting and licking elicited by each agent. ⁄:t; n= 6/group).

T. Akiyama et al. / Neuroscience 266 (2014) 38–46 41

significantly less licking (p< 0.05, t-test), was elicited by

id injection of SLIGRL-NH2 (Fig. 1B, F). Similarly, id

injection of chloroquine elicited biting and significantly

less licking (Fig. 1F) over a similar time course. In

contrast, id injection of capsaicin in the calf elicited

licking behavior that peaked 20 min post-injection, with

significantly less biting (Fig. 1C, F). Neither id injection

of saline (vehicle for histamine and SLIGRL) nor 7%

Tween-80 (vehicle for capsaicin) elicited any significant

biting or licking behavior (Fig. 1D–F, respectively).

Alloknesis and allodynia

A series of three innocuous von Frey mechanical stimuli

was applied at 5-min intervals following id chemical

injections to assess touch-evoked biting, licking or

flinching behaviors. Prior to chemical injection, von Frey

stimulation never elicited any behavioral responses.

Fig. 2A shows the time-course for touch-evoked

behaviors following id injection of histamine. Robust

touch-evoked biting (alloknesis) peaked immediately

following id injection of histamine and slowly declined

over the ensuing 45 min. The cumulative score for biting

was significantly greater (p< 0.05, unpaired t-test)following id injection of histamine (Fig. 2A) compared to

the saline control group (Fig. 2F, open bars). The very

low levels of licking and flinching did not differ from

saline controls (Fig. 2F). Following id injection of

SLIGRL-NH2 (Fig. 2B), touch-evoked biting was

Fig. 2. Time courses for alloknesis elicited by histamine and SLIGRL, and fo

of histamine (50 lg/10 ll) in calf skin, three von Frey stimuli (bending force

histamine injection. Presence or absence of biting (j), licking (4) or flinching

score (maximum value = 3). Error bars: SEM (n= 6). (B) As in A for SLIGR

Tween 80. (F) Cumulative behavior scores for von Frey stimulus-evoked biting

from saline-evoked biting (p< 0.05, unpaired t-tests, n= 6/group). $: sig

unpaired t-test, n= 6/group).

significantly less compared to histamine (p< 0.05,

unpaired t-test) but was significantly greater (p< 0.05,

unpaired t-test) compared to the saline control group

(Fig. 2F). This was accompanied by very low levels of

licking and flinching (Fig. 2B) that did not differ from

saline controls (Fig. 2F).

Following id injection of capsaicin (10 lg), there was a

rapid increase in touch-evoked flinching (allodynia) over

the initial 30 min that persisted for at least 2 h (Fig. 2C).

The cumulative score for flinching was significantly

greater (p< 0.05, unpaired t-test) following id injection

of capsaicin (Fig. 2C) compared to vehicle (7% Tween

80) controls (Fig. 2F, gray bars). The low levels of biting

and licking following id injection of capsaicin (Fig. 2C)

were not different from vehicle controls (Fig. 2F).

Electrophysiology

Recordings were made from a total of 56 units in 38 mice.

Twenty-four were classified as WDR and 32 as HT. Eight

of 22 units tested responded to id injection of histamine,

and six of 24 units tested responded to id injection of

SLIGRL-NH2. None of 10 units tested responded to id

injection of saline. All units were located in the

superficial dorsal horn at a mean depth of

137.4 lm± 11.6 below the surface. Histologically

recovered lesion sites were located in the superficial

dorsal horn (Fig. 4F).

r allodynia elicited by capsaicin. (A) At 5-min intervals after id injection

0.7 mN) were sequentially applied at separate sites 1–2 mm from the

(d) immediately after each stimulus was used to calculate a behavior

L-NH2 (50 lg/10 ll). (C) Capsaicin (10 lg/10 ll). (D) Saline. (E) 7%, licking and flinching following id injections. ⁄, #: Significantly differentnificantly different from 7% Tween-80-evoked flinching (p< 0.05,

42 T. Akiyama et al. / Neuroscience 266 (2014) 38–46

In histamine-responsive units, mean mechanically

evoked responses were significantly greater following id

injection of histamine compared to pre-injection. The

example of Fig. 3 shows an increased response to the

von Frey stimulus following id injection of histamine,

which elicited a prolonged increase in firing of this

lamina I cell. This was accompanied by an expansion of

the mechanosensitive receptive field (black area in

figurine drawings of hindpaw).

Fig. 4A plots mean peak mechanically evoked

responses vs. time for the histamine-sensitive units,

using the same von Frey stimulus that was used in the

behavioral studies (j; 0.7 mN). Mean ongoing activity is

also plotted (O, dashed line) to show increased firing

during the first 15 min post-histamine. Prior to id injection

of histamine, only 1/8 units responded to the weakest

von Frey stimulus, while all eight units responded at

5 min post-histamine. The mean mechanically evoked

response was significantly greater than the pre-injection

response (�5 in Fig. 4A) out to 45 min post-histamine

(p< 0.05; paired t-test). The failure of the 0.7 mN

stimulus to initially excite most histamine-responsive

units precluded our ability to map receptive field areas. A

stronger (55 mN) von Frey stimulus initially excited all

eight histamine-sensitive units, and elicited a significantly

greater mean response at 10 and 15 min post-histamine

(data not shown). This was accompanied by a significant

(p< 0.01, paired t-test) expansion of receptive field area

by an average of 260.4%.

Responses elicited by the same 0.7 mN von Frey

stimulus were not affected following id injection of

histamine in histamine-insensitive units (Fig. 4B), or

following id injection of SLIGRL-NH2 that either did

(Fig. 4D) or did not increase neuronal firing (Fig. 4E).

Similarly, id saline injection did not affect mechanically

Fig. 3. Enhancement of mechanically evoked response of dorsal

horn neuron after histamine. Peristimulus-time histogram (bins: 1 s)

of unit’s responses to application of von Frey filament (VF, ;,0.7 mN),) before and after id injection of histamine (;). Raw spike

traces are shown above for portions of the response (dashed lines);

dots show time of VF stimuli. Upper left inset: mechanosensitive

receptive field (blue) on hindpaw. Lower left inset: lamina I recording

site (d) on spinal section. Upper right inset: mechanosensitive

receptive field expanded following histamine.

evoked responses (Fig. 4C). In addition, there were no

consistent changes in mechanosensitive receptive field

areas under any of these treatment conditions.

DISCUSSION

Discrimination between itch and pain

The present results indicate that biting at a site of pruritic

stimulation on the lower hindlimb reflects itch, whereas

licking at a site of capsaicin injection reflects pain.

Intradermal injection of the pruritogens histamine,

SLIGRL-NH2 and chloroquine all elicited biting with little

licking, whereas id injection of capsaicin evoked licking

and little biting, consistent with a recent report (LaMotte

et al., 2011) and with the idea that hindpaw biting

reflects itch. Presumably, biting or gnawing at an itchy

site on the hindlimb serves the same antipruritic function

as scratching, since quadrupeds apparently do not

scratch an itchy hindlimb with the contralateral paw.

We interpret the hindlimb licking behavior following id

injection of capsaicin to reflect pain. Intraplantar injection

of capsaicin elicited licking behavior in rats (Klein et al.,

2011a), and hindpaw injection of formalin elicited licking

in mice (Hagiwara et al., 1999). We previously reported

that systemic administration of morphine suppressed

forelimb wiping, but not hindlimb scratching, elicited by

id cheek injections of capsaicin or mustard oil in mice

and rats, consistent with the notion that forelimb wiping

reflects pain (Akiyama et al., 2010c; Spradley et al.,

2012a). Presumably, licking of the hindpaw or forelimb

wiping directed to the cheek represent the

biomechanically available nocifensive responses to

ameliorate chemogenic pain in these two different body

areas. Licking and wiping may be akin to rubbing an

injury to relieve pain.

Alloknesis and allodynia

Following id calf injection of histamine, low-threshold

mechanical stimuli reliably elicited biting (Fig. 2A).

Neither naı̈ve mice nor control mice receiving id

injection of saline exhibited any significant touch-evoked

biting of the calf. We previously reported that following

id injection of histamine in the rostral back, low-

threshold mechanical stimuli reliably elicited hindlimb

scratch bouts over a prolonged time course in a manner

that was significantly reduced by systemic

administration of naltrexone (Akiyama et al., 2012a). We

thus interpret these touch-evoked behaviors to reflect

alloknesis. Interestingly, following id injection of

SLIGRL-NH2, which itself elicited significant biting

(Fig. 2B), low-threshold mechanical stimuli elicited a low

incidence of biting that was significantly greater than

vehicle controls, but significantly less compared to the

group receiving id injection of histamine. This is

consistent with our previous study, in which touch

stimuli failed to elicit any significant scratching following

id injection of SLIGRL-NH2 in the rostral back (Akiyama

et al., 2012a). In this latter study, reliable touch-evoked

scratching (alloknesis) was observed following id

injection of histamine, 5-HT, the PAR-4 agonist

Fig. 4. Enhancement of mechanosensitivity in histamine-sensitive dorsal horn neurons. (A) Histamine-sensitive units (n= 8). Graphs plot mean

peak responses to innocuous mechanical von Frey stimulus (0.7 mN; j, solid line), and mean unit firing rate over a 20 s period just prior to the von

Frey stimulus (O, dashed line), vs. time following id injection of histamine (at time 0). Pre-histamine responses are shown at the �5 min timepoint.

Error bars: SEM ⁄: Significant difference compared to peak response before histamine (p< 0.05; paired t-test). (B) graph as in A for 14 histamine-

insensitive units. (C) As in A for 10 units following id injection of saline (vehicle control). (D) as in A for six units responsive to id injection of SLIGRL-

NH2. (E) as in A for 18 units insensitive to SLIGRL-NH2. (F) 29 histologically recovered recording sites (dots) compiled on representative lumbar

section.

T. Akiyama et al. / Neuroscience 266 (2014) 38–46 43

AYPGKF-NH2, and the MrgprC11 agonist BAM8-22, but

not following SLIGRL-NH2 or the MrgprA3 agonist

chloroquine (Akiyama et al., 2012a). Collectively, these

data indicate that alloknesis is associated with itch

elicited by certain, but not all, pruritogens.

While both SLIGRL-NH2 and BAM8-22 are agonists

of MrgprC11 (Liu et al., 2011), it was curious that only

BAM8-22 elicited alloknesis (Akiyama et al., 2012a). We

previously reported that the histamine H1 receptor

antagonist terfenadine attenuated scratching and

alloknesis elicited by histamine, but not that elicited by

5-HT, the PAR-4 agonist or BAM8-22 (Akiyama et al.,

2012a). This is consistent with current evidence for

histamine-dependent and -independent types of itch

(Johanek et al., 2007). Separate populations of primary

afferents (Johanek et al., 2008; Namer et al., 2008) and

primate spinothalamic tract neurons (Davidson et al.,

2007, 2012) respond to histamine vs. cowhage, which

elicits itch via PAR-2 and -4 receptors (Reddy et al.,

2008). In mice, the majority of histamine-responsive

dorsal horn neurons also responded to other non-

histaminergic pruritogens such as SLIGRL-NH2,

chloroquine and 5-HT (Akiyama et al., 2009a,b, 2012b).

However, cowhage was not tested in these studies, nor

were other non-histaminergic pruritogens such as

chloroquine tested in the primate spinothalamic tract

studies. It thus remains uncertain to what extent

different histaminergic and non-histaminergic itch

mediators excite separate or overlapping populations of

spinal neurons in these two species. Recent evidence

indicates that natriuretic polypeptide B (Nppb) and

gastrin releasing peptide (GRP) are essential for

both histamine-dependent and –independent itch

transmission in mice (Sun et al., 2009; Mishra and

Hoon, 2013), implying that pruriceptors expressing

various combinations of transduction molecules (e.g.,

Mrgprs, PARs, 5-HT2, histamine H1,4, etc.) converge

onto a common population of itch-signaling spinal

neurons. However, cowhage was not tested in these

studies either, leaving open the possibility of an

additional itch-signaling pathway. More studies are

needed to address the issue of multiple itch-signaling

pathways, and why some itch mediators elicit alloknesis

while others do not.

A novel finding was that following id injection of

capsaicin, light touch reliably elicited hindlimb flinches

(Fig. 2C). We previously reported finching behavior

following intraplantar injection of capsaicin (Klein et al.,

2011b), and flinching is often used as a behavioral

readout in the formalin pain assay. We therefore

interpret touch-evoked flinching to reflect allodynia.

Sensitization of itch-signaling pathways

One potential mechanism underlying alloknesis is the

sensitization of histamine-dependent itch-signaling

neurons. We previously showed that subpopulations of

neurons in the superficial doral horn respond to id

injection of pruritogens over a time course consistent

with scratching behavior (Akiyama et al., 2009a,b),

44 T. Akiyama et al. / Neuroscience 266 (2014) 38–46

representing candidate itch-signaling neurons. In the

present study, we similarly recorded from lumbar dorsal

horn neurons with receptive fields on the hindpaw. Most

unit receptive fields were in glabrous hindpaw skin, thus

not matching exactly the site of pruritic stimulation on

furry calf skin in the behavioral studies. Despite the

drawback of an imperfect match between behavioral

and neurophysiological stimulus loci, it is advantageous

that the present neuronal population is otherwise

comparable with unit populations reported in our

previous studies (Akiyama et al., 2009a,b, 2011a,b).

The present histamine-responsive dorsal horn neurons

exhibited an enhancement of low-threshold

mechanically evoked responses following id injection of

histamine, over a time course matching that of

behavioral alloknesis. Units unresponsive to id injection

of histamine, and units tested with id injection of saline,

did not exhibit enhanced mechanically evoked

responses. Moreover, id injection of SLIGRL-NH2 had

no significant effect on low-threshold mechanically

evoked responses, regardless of whether it activated

the dorsal horn neuron or not (Fig. 4D, E). This is

consistent with the finding that the behavior score for

touch-evoked biting was significantly lower following id

injection of SLIGRL-NH2 compared to that following id

injection of histamine (Fig. 2A, B). Therefore, we believe

that these similar findings justifiy our present use of

lumbar dorsal horn units with hindpaw receptive fields

as a correlative neuronal model for the behavioral

studies conducted using furry calf skin.

In histamine-responsive units, increased

mechanosensitivity following id injection of histamine was

manifested by (a) response to the lowest-threshold

stimulus in units that did not respond to this stimulus pre-

histamine, (b) increased response to suprathreshold

mechanical stimuli following histamine, and (c) receptive

field expansion. Application of histamine sensitized

responses of visceral polymodal nociceptors to

mechanical stimulation (Koda and Mizumura, 2002),

while SLIGRL-NH2 sensitized cutaneous C-fiber

polymodal nociceptor to noxious heat but not mechanical

stimuli (Ding-Pfennigdorff et al., 2004). These data are

consistent with our observation that id injection of

histamine enhanced the responses of histamine-sensitive

dorsal horn neurons to low-threshold mechanical stimuli,

whereas such responses were not enhanced following id

injection of SLIGRL-NH2 in dorsal horn units sensitive to

this pruritogen (Fig. 4). Moreover, morphine, which

induces itch, was recently reported to enhance

pruritogen-evoked responses, as well as responses to

innocuous mechanical stimulation, of trigeminothalamic

tract neurons (Moser and Giesler, 2013). Histamine was

also reported to activate inward calcium currents in

Merkel cells (Boulais et al., 2009), and increased skin

temperature due to histamine-evoked vasodilation could

increase activity in low-threshold mechanoreceptors

(Duclaux and Kenshalo, 1972). These observations

provide potential mechanisms for histamine to

peripherally sensitize low-threshold mechanoreceptors.

Alternatively, central sensitization could account for

the enhanced low-threshold mechanosensitivity and

expanded receptive field area of histamine-sensitive

dorsal horn neurons. Many histamine-sensitive dorsal

horn neurons respond to low-threshold mechanical

stimuli, and a high percentage of these neurons also

responded to id injection of SLIGRL-NH2 (Akiyama

et al., 2009a,b). However, low-threshold mechanically

evoked responses were presently enhanced by id

injection of histamine but not SLIGRL-NH2. If central

sensitization plays a role, our findings suggest that the

dorsal horn neurons would be differentially sensitized by

input from histamine- but not SLIGRL-NH2-sensitive

primary afferents. If so, the cellular mechanism

underlying this differential sensitization remains to be

elucidated in future studies.

Alloknesis has also been observed in an experimental

model of chronic dry skin pruritus on the rostral back and

hindpaw (Akiyama et al., 2011a, 2012a). The onset of

alloknesis was delayed relative to the onset of

spontaneous scratching, and was associated with a

significant increase in Fos expression in the superficial

dorsal horn (Nojima et al., 2004). Lumbar superficial

dorsal horn neurons recorded ipsilateral to a dry-skin-

treated hindpaw exhibited significantly heightened

spontaneous activity compared to neurons ipsilateral to

the vehicle (water)-treated side, or to an untreated

hindpaw (Akiyama et al., 2011a). Curiously, in the latter

study we did not observe enhancement of mechanically

evoked responses of neurons ipsilateral to the dry skin-

treated hindpaw (Akiyama et al., 2011a). However, this

may be because alloknesis was only elicited by

stimulation at the border of the dry skin-treated area

(Akiyama et al., 2012a), while in the electrophysiological

study (Akiyama et al., 2011a) the neuronal receptive

fields were within the hindpaw dry skin area where

mechanical stimulation would not be expected to elicit

enhanced responses.

CONCLUSIONS

Intradermal injection of pruritogens including histamine,

SLIGRL-NH2 and chloroquine into the calf primarily

elicited biting behavior (with little licking), while id

injection of capsaicin elicited licking and little biting

behavior. Following id injection of histamine, innocuous

mechanical stimuli reliably elicited biting directed to the

stimulus site, interpreted as alloknesis. In contrast,

following id injection of capsaicin light touch elcited

hindlimb flinches, interpreted as allodynia. In

electrophysiological studies, histamine-responsive single

units in the superficial lumbar dorsal horn exhibited

enhanced mechanically evoked responses and

receptive field expansion, possibly reflecting peripheral

and/or central sensitization that may underlie alloknesis.

The present behavioral model thus appears to

distinguish between itch- and pain-related behavioral

responses, and provides a basis to investigate lumbar

spinal neural mechanisms underlying itch, alloknesis,

pain and allodynia.

Acknowledgments—Supported by grants DE021183 and

AR057194 from the National Institutes of Health (to E.C.) and

grant AR063228 from the National Institutes of Health (to T.A.).

T. Akiyama et al. / Neuroscience 266 (2014) 38–46 45

REFERENCES

Akiyama T, Carstens E (2013) Neural processing of itch.

Neuroscience 250:697–714.

Akiyama T, Carstens MI, Carstens E (2009a) Excitation of mouse

superficial dorsal horn neurons by histamine and/or PAR-2

agonist: potential role in itch. J Neurophysiol 102(4):2176–2183.

Akiyama T, Merrill AW, Carstens MI, Carstens E (2009b) Activation of

superficial dorsal horn neurons in the mouse by a PAR-2 agonist

and 5-HT: potential role in itch. J Neurosci 29(20):6691–6699.

Akiyama T, Carstens MI, Carstens E (2010a) Differential itch- and

pain-related behavioral responses and micro-opoid modulation in

mice. Acta Derm Venereol 90(6):575–581.

Akiyama T, Carstens MI, Carstens E (2010b) Facial injections of

pruritogens and algogens excite partly overlapping populations of

primary and second-order trigeminal neurons in mice. J

Neurophysiol 104(5):2442–2450.

Akiyama T, Carstens MI, Carstens E (2010c) Spontaneous itch in the

absence of hyperalgesia in a mouse hindpaw dry skin model.

Neurosci Lett 484(1):62–65.

Akiyama T, Carstens MI, Carstens E (2011a) Enhanced responses of

lumbar superficial dorsal horn neurons to intradermal PAR-2

agonist but not histamine in a mouse hindpaw dry skin itch model.

J Neurophysiol 105(6):2811–2817.

Akiyama T, Iodi Carstens M, Carstens E (2011b) Transmitters and

pathways mediating inhibition of spinal itch-signaling neurons by

scratching and other counterstimuli. PLoS One 6(7):e22665.

Akiyama T, Carstens MI, Ikoma A, Cevikbas F, Steinhoff M, Carstens

E (2012a) Mouse model of touch-evoked itch (alloknesis). J

Investigat Dermatol 132(7):1886–1891.

Akiyama T, Tominaga M, Carstens MI, Carstens EE (2012b) Site-

dependent and state-dependent inhibition of pruritogen-

responsive spinal neurons by scratching. Eur J Neurosci

36(3):2311–2316.

Bendsoe N, Bjornberg A, Asnes H (1987) Itching from wool fibres in

atopic dermatitis. Contact Dermat 17(1):21–22.

Bickford RG (1937) Experiments relating to the itch sensation, its

peripheral mechanism and central pathways. Clin Sci 3:377–386.

Boulais N, Pennec JP, Lebonvallet N, Pereira U, Rougier N, Dorange

G, Chesne C, Misery L (2009) Rat Merkel cells are

mechanoreceptors and osmoreceptors. PLoS One 4(11):e7759.

Carstens E, Kuraishi Y (2004) Animal models of itch: scratching away at

theproblem. In:YosipovitchG,GreavesMW,FleischerAB,McGlone

F, editors. Itch: basic mechanisms and therapy. Monticello,

NY: Marcel Dekker. p. 35–50.

Davidson S, Zhang X, Yoon CH, Khasabov SG, Simone DA, Giesler

Jr GJ (2007) The itch-producing agents histamine and cowhage

activate separate populations of primate spinothalamic tract

neurons. J Neurosci 27(37):10007–10014.

Davidson S, Zhang X, Khasabov SG, Moser HR, Honda CN, Simone

DA, Giesler Jr GJ (2012) Pruriceptive spinothalamic tract

neurons: physiological properties and projection targets in the

primate. J Neurophysiol 108(6):1711–1723.

Ding-Pfennigdorff D, Averbeck B, Michaelis M (2004) Stimulation of

PAR-2 excites and sensitizes rat cutaneous C-nociceptors to

heat. Neuroreport 15(13):2071–2075.

Duclaux R, Kenshalo DR (1972) The temperature sensitivity of the

type I slowly adapting mechanoreceptors in cats and monkeys. J

Physiol 224(3):647–664.

Graham D, Goodell H, Wolff H (1951) Neural mechanisms involved in

itch, itchy skin, and tickle sensations. J Clin Investig 30(1):37–49.

Hagiwara K, Nojima H, Kuraishi Y (1999) Serotonin-induced biting of

the hind paw is itch-related response in mice. Pain Res 14:53–59.

Heyer GR, Hornstein OP (1999) Recent studies of cutaneous

nociception in atopic and non-atopic subjects. J Dermatol

26(2):77–86.

Heyer G, Dotzer M, Diepgen TL, Handwerker HO (1997) Opiate and

H1 antagonist effects on histamine induced pruritus and

alloknesis. Pain 73(2):239–243.

Hosogi M, Schmelz M, Miyachi Y, Ikoma A (2006) Bradykinin is a

potent pruritogen in atopic dermatitis: a switch from pain to itch.

Pain 126(1–3):16–23.

Hundley JL, Carroll CL, Lang W, Snively B, Yosipovitch G, Feldman

SR, Jorizzo JL (2006) Cutaneous symptoms of dermatomyositis

significantly impact patients’ quality of life. J Am Acad Dermatol

54(2):217–220.

Ikoma A, Steinhoff M, Stander S, Yosipovitch G, Schmelz M (2006)

The neurobiology of itch. Nat Rev Neurosci 7(7):535–547.

Johanek LM, Meyer RA, Hartke T, Hobelmann JG, Maine DN,

LaMotte RH, Ringkamp M (2007) Psychophysical and

physiological evidence for parallel afferent pathways mediating

the sensation of itch. J Neurosci 27(28):7490–7497.

Johanek LM, Meyer RA, Friedman RM, Greenquist KW, Shim B,

Borzan J, Hartke T, LaMotte RH, Ringkamp M (2008) A role for

polymodal C-fiber afferents in nonhistaminergic itch. J Neurosci

28(30):7659–7669.

Klein A, Carstens MI, Carstens E (2011a) Facial injections of

pruritogens or algogens elicit distinct behavior responses in rats

and excite overlapping populations of primary sensory and

trigeminal subnucleus caudalis neurons. J Neurophysiol

106(3):1078–1088.

Klein AH, Sawyer CM, Zanotto KL, Ivanov MA, Cheung S, Carstens

MI, Furrer S, Simons CT, Slack JP, Carstens E (2011b) A tingling

sanshool derivative excites primary sensory neurons and elicits

nocifensive behavior in rats. J Neurophysiol 105(4):1701–1710.

Koda H, Mizumura K (2002) Sensitization to mechanical stimulation

by inflammatory mediators and by mild burn in canine visceral

nociceptors in vitro. J Neurophysiol 87(4):2043–2051.

LaMotte RH, Shimada SG, Sikand P (2011) Mouse models

of acute, chemical itch and pain in humans. Exp Dermatol

20(10):778–782.

Liu Q, Weng HJ, Patel KN, Tang Z, Bai H, Steinhoff M, Dong X (2011)

The distinct roles of two GPCRs, MrgprC11 and PAR2, in itch and

hyperalgesia. Sci Signal 4(181):ra45.

Mason R (2008) Fabrics for atopic dermatitis. J Family Health Care

18(2):63–65.

Mishra SK, Hoon MA (2013) The cells and circuitry for itch responses

in mice. Science 340(6135):968–971.

Moser HR, Giesler Jr GJ (2013) Itch and analgesia resulting from

intrathecal application of morphine: contrasting effects on different

populations of trigeminothalamic tract neurons. J Neurosci

33(14):6093–6101.

Namer B, Carr R, Johanek LM, Schmelz M, Handwerker HO,

Ringkamp M (2008) Separate peripheral pathways for pruritus in

man. J Neurophysiol 100(4):2062–2069.

Nojima H, Cuellar JM, Simons CT, Carstens MI, Carstens E (2004)

Spinal c-fos expression associated with spontaneous biting in a

mouse model of dry skin pruritus. Neurosci Lett 361(1–3):79–82.

Reddy VB, Iuga AO, Shimada SG, LaMotte RH, Lerner EA (2008)

Cowhage-evoked itch is mediated by a novel cysteine protease: a

ligand of protease-activated receptors. J Neurosci

28(17):4331–4335.

Ricci G, Patrizi A, Bellini F, Medri M (2006) Use of textiles in atopic

dermatitis: care of atopic dermatitis. Curr Prob Dermatol

33:127–143.

Shimada SG, LaMotte RH (2008) Behavioral differentiation between

itch and pain in mouse. Pain 139(3):681–687.

Simone DA, Alreja M, LaMotte RH (1991) Psychophysical studies of

the itch sensation and itchy skin (‘‘alloknesis’’) produced by

intracutaneous injection of histamine. Somatosens Motor Res

8(3):271–279.

Spradley JM, Davoodi A, Carstens MI, Carstens E (2012a) Opioid

modulation of facial itch- and pain-related responses and

grooming behavior in rats. Acta Derm Venereol 92(5):515–520.

Spradley JM, Davoodi A, Gee LB, Carstens MI, Carstens E (2012b)

Differences in peripheral endocannabinoid modulation of

scratching behavior in facial vs. spinally-innervated skin.

Neuropharmacol 63(4):743–749.

46 T. Akiyama et al. / Neuroscience 266 (2014) 38–46

Sun YG, Zhao ZQ, Meng XL, Yin J, Liu XY, Chen ZF (2009) Cellular

basis of itch sensation. Science 325(5947):1531–1534.

Wahlgren CF (1991) Itch and atopic dermatitis: clinical and

experimental studies. Acta Derm Venereol Suppl 165:1–53.

Wahlgren CF (1999) Itch and atopic dermatitis: an overview.

J Dermatol 26(11):770–779.

(Accepted 4 February 2014)(Available online 12 February 2014)